Natural-language creation
Describe the audience, workflow, app behavior, analytics, and boundaries without writing code.
LearnAdapt Agentic Studio
Build plugins with natural language
No coding needed. Your prompt becomes the client brief for an agent development team that plans, designs, builds, reviews, validates, and prepares an installable LearnAdapt app for preview and admin approval.
Agentic Studio FAQ
Do I need to code?
No. You can describe the app or learning experience in natural language. Advanced users can add implementation constraints, data boundaries, platform capabilities, API hooks, and validation expectations.
Will the generated plugin behave like a real app?
That is the goal of Agentic Studio. Approved plugins are packaged as LearnAdapt apps. After directory approval and user installation, they register in the selected LearnAdapt surface, such as My Apps, a course space, student activity, teacher dashboard, analytics dashboard, or admin tools.
What happens before the agents start building?
A deterministic preflight helper checks the brief against the current PedOS SDK contract, warns about missing details, records unsupported requests as future PedOS KIV, and confirms that the account has enough credits to start the build.
What kind of data should the app save?
Studio infers this from the brief. Workflow tools should save lightweight app activity. Scored learning tasks should save learning progress. Adaptive support workflows should save the support choices needed for review. Research fields require study mode and consent.
When does app activity save?
Preview activity is local or sandboxed. Persistent LearnAdapt activity is accepted only after admin approval, directory deployment, and user installation. Research-study fields are accepted only when study mode and participant consent are active.
What is learning progress data?
Learning progress data captures the minimum signals needed to support useful analytics: the skill or concept, the task, whether the learner succeeded, time or attempt order, and response time. Optional fields can include confidence, hints, difficulty, and misconception tags.
What is adaptive support data?
Adaptive support data adds context for apps that provide hints, route support, or adjust the next step. It records enough to explain what help was given and why, without showing technical traces to learners or teachers.
How does admin approval work?
Submitted plugins run package validation, static checks, sandbox checks, data-boundary review, route health checks, and admin preview. Only approved plugins can appear in the Plugins Directory for users to install.
Can workflow plugins avoid learning-data collection?
Yes. Timers, lesson planners, rotation boards, and other workflow apps should save only the activity needed to run the workflow. They should not collect correctness, score, mastery, or adaptive-support data unless the creator explicitly asks for assessed learning progress.
For authenticated LearnAdapt users
Turn a serious learning idea into an installable LearnAdapt app.
Agentic Studio is not a throwaway demo. It is the authenticated build environment where your natural-language instructions become the brief for an agent development team that creates apps for LearnAdapt runtime surfaces, approved data, storage, analytics, and admin-review boundaries.
Metered AI features use the same LearnAdapt account credits as the rest of the platform. Generation starts only after authentication so usage, credits, ownership, and review history remain traceable.
Real LearnAdapt apps
Approved plugins open in their declared LearnAdapt surface, use scoped storage and settings, and behave like installed apps.
Admin-reviewed distribution
Package, runtime, data-boundary, sandbox, and route checks run before directory approval.
Data by boundary
Apps collect only approved app activity, learning progress, adaptive support, or research-study data.
0. PreflightCheck the brief against the current SDK, flag future-version requests, and estimate account credits before generation.
1. BuildDescribe the app in natural language; Studio turns the brief into a LearnAdapt app package.
2. PreviewOpen the generated app in a sandbox, test its UI, settings, storage, and runtime behavior.
3. IterateAsk Studio to refine screens, workflows, analytics, data boundaries, or guardrails before submission.
4. SubmitSend the plugin package, icon, data boundary, and review notes to the directory queue.
5. Admin approveAdmins validate package safety, route health, data scope, and installed-app behavior.
6. InstallApproved plugins appear in the Plugins Directory so users can install them into the declared runtime surface.
7. Open in LearnAdaptInstalled plugins launch inside LearnAdapt with user-scoped settings, storage, analytics, and lifecycle hooks.
8. EvidenceApproved app activity writes to LearnAdapt evidence stores for analytics and research exports.